HKUDS Releases DeepCode v2 Open-Source Coding Agent
HKUDS has released DeepCode v2, an open-source coding agent designed for real-world software engineering with continuous loop engineering and verified delivery.
Read more →
Momentic Unveils Agentic Testing Platform to Fix AI-Era QA
Momentic announces a major platform update and a $15M Series A to replace flaky end-to-end test scripts with autonomous, self-healing AI agents.
Read more →
Hermes Agent Now Runs Natively on Windows — No WSL, No Docker Required
Nous Research has announced native Windows support for Hermes Agent — the open-source AI agent OS. Windows 10 and Windows 11 users can now install and run Hermes without WSL, Cygwin, or Docker, getting access to the full CLI, TUI, browser tool, MCP server support, and 15+ messaging platform integrations.
Read more →
Mobilewright: Playwright for Real Mobile Devices — Deterministic AI Agent Access to iOS and Android
Mobile Next has launched Mobilewright — Playwright for real mobile apps and devices. It gives AI agents deterministic, flake-free access to iOS and Android through one unified API, supporting real devices, simulators, and emulators. It's the programmable mobile runtime built for the AI era.
Read more →
Google Antigravity CLI: Bring the Agent-First IDE to Your Terminal
Google Antigravity has launched a lightweight CLI that brings its agentic development platform directly to the terminal — same harness, same models, fully adapted to your keybindings, themes, and workflows. No IDE required.
Read more →
Cue: The macOS App That Brings AI to Every Text Field
Cue is a 2.5MB macOS app that brings AI into every text field on your Mac with a double-tap of the right Command key — no tab switching, no copy-pasting, no losing your thread. Works in Slack, iMessage, Gmail, Linear, Notes, Terminal, and anywhere else a cursor blinks.
Read more →
Mercury: The Open-Source AI Agent That Thinks, Acts, and Asks Permission
Mercury is an open-source AI agent built around three principles: it thinks with intelligent loop detection and 25-step agentic loops, acts with 40+ built-in tools across filesystem, shell, GitHub, and Spotify, and asks — ensuring nothing happens without your explicit permission or AI self-monitoring.
Read more →
Agent Zero v1.16 + A0 CLI v1.10: Computer Use, Modular Speech, and Plugin Overhaul
Agent Zero v1.16 and A0 CLI v1.10 are out, bringing Computer Use as a cross-platform beta feature, a fully modular speech stack with bring-your-own STT/TTS plugins, configurable timezone preferences, ephemeral screenshot privacy, and a major overhaul of the plugin and editor system.
Read more →
GBrain: Garry Tan's Open-Source AI Agent Brain With Self-Wiring Knowledge Graph
GBrain is an open-source AI agent memory system built by Garry Tan, President and CEO of Y Combinator, to power his personal agents in production. It features a self-wiring typed knowledge graph, hybrid vector + graph search with P@5 of 49.1%, and a fully autonomous doctor command that optimizes brain health without human intervention.
Read more →
NVIDIA Launches Verified Agent Skills: Transparency and Security for Agentic AI Pipelines
NVIDIA has released Verified Agent Skills — a system for adding transparency, provenance, and risk metadata to AI agent skills. Each verified skill carries a skill card detailing what it does, where it came from, and whether it's been modified, built on the open agentskills.io specification and compatible with Claude Code, OpenAI Codex, and Cursor.
Read more →